Integrate already built ML classifier with a chatbot

I want to integrate a multi-class classification model of diseases with a rasa chatbot. I expect the bot to ask about symptoms and then it predicts the disease. I already built the model in Jubyter notebook but I need a tutorial or already built chatbot similar to what I want if you can help, please? Thanks in advance

Hey @nadamakram, welcome to the forum :blush:.

You can Check this video by @rctatman,


To summerize my approach:

  • save encoders & model with joblib
  • load into new python environment
  • use encoders to encode input, predict with saved model, and reverse encoding

I just added all the relevant code to my github repo if you want to take a look. (I don’t have permission to share the data, so the training won’t work but you should be able to load the saved model.)

Let me know if you have any other quesions! :slight_smile: